Multidisciplinary and Multiobjective Optimization in Conceptual Design for Mixed-Stream Turbofan Engines

摘要

Despite major advances in design tools such as engine cycle analysis software andcomputer aided design, conceptual gas turbine engine design is essentially a trial-and-error process based on the experience of engineers. Modern optimization concepts, such as multidisciplinary optimization (MDO), and multiobjective optimization (MOO), linked with sequential quadratic programming (SQP) methods and genetic algorithms (GA), were applied to the conceptual engine design process to automate the conceptual design phase. Robust integrated computer codes were created to find the optimal values of eight engine parameters in order to minimize fuel usage, aircraft cost and engine annulus area over a given mission. The engine cycle selected for study was the mixed stream, low bypass turbofan.